1 / 17

Understanding PV Access in EPICS 7: Python Examples

Gain insights into PV Access in EPICS 7 through Python examples presented by Amanda Carpenter on History Channel Access. Explore numbers, enums, strings, scalars, arrays, alarms, limits, and more. Learn about PV Data structures like EPICS V4, normative types, and custom data handling. Discover tools such as cainfo, caget, and camonitor for Channel Access vs. PV Access comparison. Experiment with custom PV Data in CS-Studio using PVA protocol. Python examples demonstrate how to use PV Access API as a provider and interact with custom data in PV Access. Dive into Python servers and clients to handle PV Data seamlessly. Explore examples for Display Builder, table server, and more to enhance your understanding.

scholz
Download Presentation

Understanding PV Access in EPICS 7: Python Examples

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. PV Access Oct. 2018 Kay Kasemir Python examples by Amanda Carpenter

  2. History Channel Access • DBR_*: Numbers, enums, string, scalar and array, with time, alarm, limits • Since beginning of EPICS • Still fully supported PV Access • PV Data: Arbitrary structures • Started as “EPICS V4” development • Since EPICS 7 (Dec. 2017) included in EPICS base

  3. PV Access Fundamentally similar to Channel Access • Name search via UDP • Connection for data transfer via TCP • EPICS_PVA_ADDR_LIST, EPICS_PVA_AUTO_ADDR_LIST Get, put, monitor • Plus an ‘RPC’ type operation Arbitrary PV Data structures instead of DBR_.. types

  4. Custom Data: Great, but then what? structure:short leveldouble datastring  typetime  stamp… structure:short leveldouble wertstring  typlong  zeit… structure:double valueshort   statusshort   severitystring  unitstime  timeStamp… • Which number to show on a user display? • What units? • Is this an alarm? • Time stamp? structure:short infodouble contentstring  metalong  ms…

  5. “Normative Types” • PV Access epics:nt/NTScalar:double valueshort   statusshort   severitystring  unitstime  timeStamp… You get what you request(but network only transfers changes) • Channel Access struct  dbr_ctrl_double:short   statusshort   severityshort    precisionchar    units[8]… no timestamp …double  valuestruct  dbr_time_double:short   statusshort   severitytimestamp stampdouble  value  You get what you request(network always transfers complete struct)

  6. Channel Access vs. PV Access EPICS 7 IOCs include PVA serverSimilar command line tools: cainfo training:ai1 caget training:ai1 camonitor training:ai1 caget –d CTRL_DOUBLE training:ai caget training:ai1.SCAN pvinfo training:ai1 pvget training:ai1 pvget –m training:ai1 pvget -r 'field()' training:ai1 pvget training:ai1.SCAN

  7. CS-Studio: Use ‘pva://…’

  8. Images: Normative type NTNDArray • Served by Area Detector (NDPluginPVA) or ‘start_imagedemo’ • pvinfo IMAGE • Value, dimensions, codec • CS-Studio: Image widget • Only needs pva://IMAGE

  9. Custom PV Data SNS Beam Lines started to use this in ~2014 start_neutrondemo pvinfo neutrons Allows fetching just what’s needed: # For detector pixel displaypvget -r 'field(pixel)' neutronspvget –m -r 'field(timeStamp, pixel)' neutrons # For energy displayspvget –m -r 'field(time_of_flight, pixel)' neutrons

  10. Custom PV Data in CS-Studio Cannot handle arbitrary structure pva://neutrons Can handle fields which arescalar or array pva://neutrons/proton_charge pva://neutrons/pixel

  11. PV Access and Python Basic ‘get’ cd ~/epics-train/examples/python/ python example1.py ‘monitor’ python example2.py

  12. PV Access API with Channel Access as “Provider” PV Access supports both the actual PvAccess protocol but also Channel Access. New tools, written for PVA, can thus fall back to CA: python example3.py Tools like CS-Studio can use both ca:// and pva://, so multiple transition options.

  13. Custom PV Data in Python Client Python receives data as dictionary, access to any element python neutrons.py

  14. Custom PV Data from Python Server Surprisingly easy: # Server python server.py # Client pvinfo pair pvget -m -r "x, y" pair pv = PvObject({'x': INT, 'y' : INT}) server = PvaServer('pair', pv) x = 1 while True: pv['x'] = x pv['y'] = 2*x server.update(pv)     sleep(1)     x = x + 1

  15. More Examples Display Builder pva_server_ramp Python code that serves ‘pva://ramp’ with alarm, prec, timestamp, … Display Builder table_server Python code that serves ‘pva://table’ as “NTTable” • Not practical to replace regular IOCs with python,but useful when custom data is needed

  16. Custom PV Data from IOC Records `makeBaseApp.pl –t example` includes “group”, see ~/epics-train/examples/ExampleApp/Db/circle.db Calc records ..:circle:x & ..:circle:y compute (x, y) coordinate on circle info() annotations create PV ”training:circle” PV as struct { angle, x, y } PVA “training:circle” updates atomically camonitortraining:xtraining:y receives separate x, y updatespvget –m training:circlewill always see sqrt(x2+y2)==1 cd ~/epics-train/examples/pythonpython circle.py

  17. PV Access • Update to Channel Access • Both can be used in parallel • Similar, but supports custom data types • Won’t replace IOC, but useful for special cases • Since EPICS 7 included in base • Details of ‘group’, PVA gateway, ‘field(…)’ access still evolving

More Related